Due to traumatic brain injury (TBI), numerous studies have focused on comprehensively determining the mechanical properties of the brain. This study examined the strain rate dependence of porcine brain under compression, and the microstructural damage was quantified using a confocal microscope and graphical user interface (GUI). The selected strain rates were 0.10 s-1, 0.025 s-1, and 0.00625 s-1 while the strain levels targeted for confocal imaging were 15%, 30%, and 40%. This study also characterized the stress-state dependence at a strain rate and strain level of 0.10 s-1 and 40%, respectively, under compression, tension, and shear. Strain rate dependency data exhibited viscoelastic behavior, and the analysis parameters correlated with increasing strain rate and strain level. Stress-state dependency data demonstrated distinct nonlinear behavior, and disparities were observed in the analysis parameters between different testing modes. Finite element procedures can implement this supplementary data for devising more realistic models.

No description available.

This dissertation is composed of three essays that analyze the significance of the Food Stamp Program (FSP) for low-income households. The first essay entitled “Intensity of Food Stamp Use and Transient and Chronic Poverty: Evidence from the Panel Study of Income Dynamics” examines the impact of intensity of use of FSP benefits on household exposure to transient and chronic poverty with respect to food and housing expenditures. The study finds that FSP is used for both long-term expenditure support and as a smoothing mechanism before the welfare reform, and only for smoothing expenditures after the welfare reform. Factors that influence both components of poverty are number of children, human capital, minority status and local economic conditions. Another finding is that shorter recertification periods reduce the length of FSP use, and indirectly result in higher poverty. The second essay entitled “The End of the Paper Era in the Food Stamp Program: The Impact of Electronic Benefits on Program Participation” documents the impact of the implementation of the statewide Electronic Benefits Transfer (EBT) system on household participation behavior in the entire period of nationwide implementation. The major finding is that the switch from paper coupons to EBT cards induces participation among eligible households, most likely by reducing the stigma associated with FSP participation. The effect of the EBT system on participation probabilities is the largest among households residing in the rural South, those not headed by a single mother or those with a White household head. The third essay entitled “The Dynamics of Food Stamp Program Participation: A Lagged Dependent Variable Approach” investigates the existence of state dependence and its sources by analyzing the dynamics of participation in the FSP using a lagged dependent variable approach. Results show that FSP receipt in the previous period is an important determinant of current FSP receipt. Estimated persistence rates declined significantly after 1996, suggesting that long-term welfare dependency was reduced after the welfare reform, at least with respect to the FSP. The source of state dependence in FSP participation among low-income households is mostly structural implying that a welfare trap does exist for these households. / Ph. D.

In this paper, I examine the existence and roles of state dependence and switching costs in the mass transition from MySpace to Facebook during the 2007-2008 time period. Using a dataset that compiles individual browsing behavior and a discrete multinomial logit model, I find precise, yet extremely small amounts of state dependence for users of only MySpace, of only Facebook, and users of both MySpace and Facebook. Positive state dependence directly implies the existence of switching costs for each firm. While there is an abundance of literature regarding switching costs in the brick and mortar setting with tangible products and services, my work extends similar analysis to the relatively new online, social networking industry by studying consumer usage trends of the two networking powerhouses: MySpace and Facebook.

Poverty analysis has been confined to incidence studies in Turkey. In the last decade research has focused on poverty persistence referred to as &bdquo / new poverty

This thesis presents three chapters on the subject of unemployment persistence. Two of the chapters are empirically focussed and the other is a purely theoretic work. Unemployment persistence is defined as the existence of serial correlation in individual employment outcomes. The first chapter finds evidence for unemployment persistence among men and women in the Australian youth labour market. Individual labour market dynamics are analysed using the Australian Longitudinal Survey. The analytic framework used is a Random Effects Probit model, incorporating lagged employment status as an explanatory variable status. Results support a “scarring” effect of unemployment upon individuals’ future employment prospects. The second chapter provides decision-theoretic foundations for unemployment persistence, based upon heterogeneous intrinsic productivity among workers. A representative firm is assumed to receive an imperfectly precise signal of worker ability every period, and re-forms its beliefs every period using a Bayesian updating method. A model of the dynamic behaviour of optimal employment decisions by the firm is constructed. It is shown that under certain circumstances workers of all productivities may be “scarred” in the eyes of the firm by past unemployment, due to the firm’s being unwilling to hire from an unemployment pool of dubious quality. The third chapter presents a detailed investigation into how to measure unemployment persistence within the UK. The chapter presents several modelling strategies capable of being used to analyse panel data of a binary nature, and discusses how to decide which methods are most appropriate in particular environments. Panel data on men from the British Household Panel Survey are used to estimate a structural state dependence equation in employment status, where lagged employment status is used as an explanatory variable. Particular attention is given to controlling for unobserved heterogeneity between individuals. The empirical results indicate strong evidence of unemployment persistence.

Cette thèse se compose de quatre travaux indépendants. Le premier concerne les modèles partiellement identifiés, c'est-à-dire des modèles dans lesquels la valeur du paramètre d’intérêt ne peut pas être déduite de la distribution des données et des hypothèses du modèle. Dans certaines situations, aucune ou au contraire plusieurs valeurs du paramètre d’intérêt sont compatibles avec les données et les hypothèses du modèle. Ce travail démontre entre autre que si l’ensemble des distributions de probabilités compatibles avec le modèle est convexe, alors les parties extrêmes de ce convexe caractérise l’ensemble des distributions compatibles avec le modèle. Le deuxième travail propose une méthode basée sur une condition d’exclusion pour corriger de l’attrition endogène dans les panels. Nous appliquons cette méthode pour estimer les transitions sur le marché du travail à partir de l’enquête emploi française. Le troisième travail propose une méthode simple pour estimer un modèle logistique avec effets fixes et dépendance d’état tel qu’étudié par Honoré et Kiriazidou. Il propose également un nouvel estimateur des écarts-types qui semble avoir de meilleures propriétés à distance finie. Le quatrième travail est une évaluation sur les collèges de la politique éducative des Réseaux-Ambition-Réussite lancée en 2006. Nous exploitons une discontinuité dans la sélection des collèges pour comparer entre eux certains collèges « identiques » avant la mise en place de la politique. Les résultats de cette évaluation laissent place à peu d’optimisme concernant l’efficacité de cette politique. / This work consists of four independent chapters. The first deals with partially identified models, i.e. models in which the parameter of interest is not uniquely identified from the data and the model’s assumptions. In some cases, no value of the parameter of interest is compatible with the data and the model’s assumption. And conversely in some cases, many values are compatible with the data and the model’s assumptions. This work demonstrates among other things that if the set of probability distributions compatible with the model is convex, then this set is characterized by its the extreme parts. The second paper proposes a method based on an exclusion restriction to correct endogenous attrition in panels. We apply this method to estimate the transitions on the labor market from the French Labour Force Survey. The third paper proposes a simple method to estimate a Logit model with fixed effects and state dependence as studied by Honoré and Kiriazidou. It also proposes a new method for inference. The fourth work is an evaluation of a French educational policy, implemented in 2006 whose objective is to provide extra resource to a subset of junior high schools. We exploit a discontinuity in treatment selection to build a credible counterfactual. Our results indicate some adverse treatment effects.

Paper [I] studies equality of opportunity in Sweden. The distinction between circumstances that constrain an individual’s opportunities and the individual choices also affecting a particular outcome is the main idea of theories of equality of opportunity. In this study, equality of opportunity is analyzed for Swedish data using a large set of variables indicating different circumstances likely to affect an individual’s opportunities. A semiparametric model is estimated to allow for a possible nonlinear relation between parental income and the income of the adult child. The reason is a hypothesis that a constrained investment behavior would make the relationship nonlinear. The results indicate significant inequality of opportunities. However, they do not indicate a nonlinear relationship between parental income and the income of the adult child. Thus, the hypothesis that low income families will have a constrained investment behavior in human capital formation is brought into question as the explanation of intergenerational income correlation in Sweden. Paper [II] focuses on the persistence of poverty in Sweden. The purpose is to distinguish between two different reasons why poverty could persist on an individual level. By using a sample of identical twins, this study takes advantage of the similarity within pairs of twins to separate family specific heterogeneity from true state dependence, where the experience of poverty leads to a higher risk of future poverty. The results, based on a four variate probit model, show the importance of true state dependence in poverty. When using the information on whether an individual received social assistance as a measure of poverty, family specific heterogeneity explains between 24 and 31 percent of the poverty persistence in the sample. Paper [III] analyzes the consequences of unemployment for a Swedish sample of couples. The purpose is to estimate the possible income replacement that a spouse can provide. Unemployment can also affect the probability that the couples split up. Since not all couples remain in the analysis, a potential selection problem can occur. To deal with this problem, and also to take care of unobserved heterogeneity, a sample selection model for panel data is estimated. The results indicate that it is necessary to take into account the selection problem. A period in unemployment is found to be correlated with a higher female income only in the case of men who earned a fairly high income before becoming unemployed. Women who earned a fairly low income and were subject to a long period of unemployment are found to be compensated by a higher male income.

The polyamines, putrescine, spermidine and spermine, are present in high concentrations in the central nervous system and, because of their policationic nature, they can interact with diverse cellular anionic targets (nucleic acids and proteins) and modulate the learning and memory by interacting with the polyamine binding site at N-methyl-D-aspartate receptor. In this study we investigated the effects of the systemic administration of spermidine and arcaine on the memory of the inhibitory avoidance task in rats. It was also determined whether the effects of the spermidine and arcaine involve state dependence. The animals were trained in an inhibitory avoidance apparatus (0.4 mA, 3s) and tested in the same apparatus 24 hours later. Immediate post-training administration of spermidine (50 mg/kg, i.p.) improved, while arcaine (10 and 30 mg/kg, i.p.) impaired step-down latencies in the inhibitory avoidance test. Administration of spermidine (50 mg/kg, i.p.) 15 minutes before testing did not alter the performance of rats which were injected with spermidine (50 mg/kg, i.p.) or vehicle immediately after training. However, administration of arcaine (30 mg/kg, i.p.) 15 minutes before testing reverted the impairment of memory caused by the administration of arcaine (30 mg/kg, i.p.) immediately after training. Administration of arcaine (30 mg/kg, i.p.) or MK-801 (0.03 mg/kg, i.p.) before testing partially reverted the impairment of memory caused by the administration of arcaine (30 mg/kg, i.p.) or MK-801 (0.03 mg/kg, i.p.) immediately after training, characterizing a crossed state dependence. The results suggest that memory improvement caused by the administration of spermidine immediately after training is not due to state dependence. In contrast, the impairment of memory induced by arcaine is due to state dependence. The crossed state dependence between arcaine and MK-801 supports that state dependence induced by arcaine is related to NMDA receptor hypofunction. / As poliaminas, putrescina, espermidina e espermina estão presentes em altas concentrações no sistema nervoso central e, por sua natureza policatiônica, podem interagir com sítios aniônicos de macromoléculas (ácidos nucléicos e proteínas) e modulam o aprendizado e a memória interagindo com o sítio de ligação das poliaminas no receptor N-metil-D-apartato. Neste estudo nós investigamos os efeitos da administração sistêmica de espermidina e arcaína sobre a memória da tarefa de esquiva inibitória em ratos. Também foi determinado se os efeitos da espermidina e arcaína envolvem dependência de estado. Os animais foram treinados em um aparelho de esquiva inibitória (0,4 mA, 3 seg) e testados no mesmo aparelho, 24 horas depois. A administração imediatamente após o treino de espermidina (50 mg/kg, i.p.) melhorou, enquanto que arcaína (10 e 30 mg/kg, i.p.) prejudicou a latência de descida da plataforma no teste da esquiva inibitória. A administração de espermidina (50 mg/kg, i.p.) 15 minutos antes do teste não afetou a performance dos ratos que foram injetados com espermidina (50 mg/kg, i.p.) ou veículo imediatamente após o treino. Entretanto, a administração de arcaína (30 mg/kg, i.p.) 15 minutos antes do teste reverteu o prejuízo da memória causado pela administração de arcaína (30 mg/kg, i.p.) imediatamente após o treino. A administração de arcaína (30 mg/kg, i.p.) ou MK-801 (0,03 mg/kg, i.p.) antes do teste reverteu parcialmente o prejuízo da memória causado pela administração de arcaína (30 mg/kg) ou MK-801 (0,03 mg/kg) imediatamente após o treino, caracterizando dependência de estado cruzada. Estes resultados sugerem que a melhora da memória causada pela administração de espermidina imediatamente após o treino não é devido à dependência de estado. Em contraste, o prejuízo da memória induzido pela arcaína é devido a uma dependência de estado. A dependência de estado cruzada entre arcaína e MK-801. sugere que a dependência de estado induzida pela arcaína envolve a hipofunção do receptor NMDA.

This dissertation consists of three articles on social assistance benefit receipt dynamics in European countries. The first article presents an analysis of state dependence in benefit receipt in Germany based on annual survey data from the German Socio-Economic Panel. The observation period extends from 1995 to 2011, thus covering the 2005 'Hartz reforms'. I estimate a series of dynamic random-effects probit models to control for observed and unobserved heterogeneity and the endogeneity of initial conditions. The high observed state dependence has a substantial structural component, with benefit receipt one year ago being associated with an increase in the likelihood of receipt today by 13 percentage points. There is only little evidence for time-variation in state dependence. The second article presents evidence on spell durations and the frequency of repeat spells using monthly administrative data from Luxembourg, the Netherlands, Norway and Sweden. In the two Nordic countries, short-term benefit receipt is the norm, with only around 6% and 11% of spells in Norway and Sweden lasting longer than 12 months. Most recipients however have multiple spells. In Luxembourg and the Netherlands, long-term benefit receipt is frequent, with median spell durations of 14 and 9 months, respectively, and one-third and one-quarter of all spells lasting 24 months or longer. The total duration of benefit receipt across spells is much higher in the Netherlands and Luxembourg than in Norway and Sweden. The third article tests the validity of one of the central assumptions of dynamic discrete-choice models of benefit dynamics, the conditional Markov property. Using monthly administrative data from Norway, the article shows that the Markov property is violated as estimated state dependence is affected by the chosen time unit of analysis. The standard model can be improved by permitting for different entry and persistence equations and duration and occurrence dependence in benefit receipt.